Chapter 3 Design from an IoT application for the organization. In the layer, you can use TensorFlow, Apache OpenNLP, Apache Tika, and other ML and NLP libraries for sentiment analysis, image analysis, document analysis, time-services analysis, and other processing. IoT data can help recognize trends common in a machine that help us to understand the breakdown of applications. At each layer in the design, you can run various ML libraries as needed. In Figure 3- 23, IoT Integration has four open-source software components designed to collect the data from external environment (i.e., sensors and boards); data stores (i.e., NoSQL databases); management (i.e., integration of IoT framework and platform), which provides the communication to collect data from a group of sensors from multiple locations, visualization (UI applications), and manipulation (ML) of time series IoT data in an easy and scalable manner. IoT frameworks like ARIoT, AllJoyn, Iotivity, etc. have implemented SOA, which is beneficial in IoT integration. These frameworks provide data interoperable capabilities, the API layer. Physically sensed activities generate the events, and these events are sent via the communication protocol such as MQTT to the service (middleware) layer. These events are pushed via the event bus to a REST API and then the update is reflected on web portals and mobiles devices. IoT Case Study Integration of technology such as AI/ML and big data along with blockchain will prove to be extremely useful for IoT use cases in the near future. The following use cases will help you to understand the IoT applications and implementation so you are able to implement IoT-based applications in your organization. • Asset tracking: Logistics organization already has tracking assets using IoT. Organizations can have real-time access to the appropriate location of the assets. You can attach tagging sensors with the assets that will improve efficiency, as it will resolve the problem of locating the assets. IoT-enabled things assist people to improve productivity. • Smart cities and real-time streaming data: Smart cities are based on connected technology that makes cities more progressive and have data to get insight that can help in improving safety, economy, and quality of life. Real-time data streaming through streaming 95